Cyber-security system and method for detecting escalation of privileges within an access token

According to one embodiment, a method detecting and mitigating a privilege escalation attack on an electronic device is described. The method involves operations by a user agent mode operating within a user space and a kernel driver mode operating within a kernel space. The kernel driver mode, in response to detecting an initial activation of a process being monitored, stores metadata associated with an access token. This metadata includes the initial token state information. Responsive to detecting an event associated with the process being monitored, the kernel mode driver extracts a portion of current state information for the access token for comparison to a portion of the stored token state information. Differences between content within the current state information and the stored token state information are used, at least in part, by the user agent mode to detect a privilege escalation attack.

Skip to: Description  ·  Claims  ·  References Cited  · Patent History  ·  Patent History
Description
CROSS REFERENCE TO RELATED APPLICATIONS

This application is based upon and claims the benefit of priority from U.S. Provisional Patent Application No. 62/690,835 filed Jun. 27, 2018, the entire contents of which are incorporated herein by reference.

FIELD

Embodiments of the disclosure relate to the field of cybersecurity. More specifically, one embodiment of the disclosure relates to a token analysis system to detect privilege escalation that is symptomatic of a cybersecurity attack.

GENERAL BACKGROUND

Cybersecurity attacks have become a pervasive problem for organizations as many electronic devices and other resources have been subjected to attack and compromise. A cybersecurity attack, also referred to as a “cyberattack,” may involve the infiltration of malicious software (e.g., malware) onto an electronic device, followed by the malware attempting to discretely gain access to sensitive information from a data store within the infected electronic device or within a resource accessible via the infected electronic device. Thereafter, the malware may attempt to alter, destroy, exfiltrate or render non-accessible the sensitive information from the infected electronic device or resource unbeknownst to the entity in control of that electronic device or resource.

Some types of cyberattacks, namely privilege escalation attacks, have become increasingly common and have led to the loss of sensitive data and compromises of electronic devices. In a privilege escalation attack, malware authors may subvert the use of an otherwise legitimate application or binary, running with proper privilege settings, by surreptitiously escalating (e.g., increasing) privileges within an access token. In many situations, privilege escalation attacks use a variation of token stealing, or token manipulation, to escalate the privileges within the access token. After obtaining escalated privileges, the malware may conduct nefarious actions that the electronic device would not be permitted without those privileges.

Herein, an “access token” is an object that contains a set of privileges (e.g., one or more privileges) that control access by a user to a resource available to an electronic device (e.g., endpoint, server, etc.) and control whether an instruction (or operation) may be executed. For certain types of resources, such as components managed by an operating system (OS), access controls may be applied where the OS ensuring that only authorized processes can utilize the resources. In response to an attempt by the user to access the resource, a portion of the OS deployed within the electronic device determines whether the user is permitted access to the resource. This determination is conducted by accessing content from the access token accompanying a resource request to determine (i) whether the user possesses the necessary privileges to access the resource and (ii) what degree or level of access to content maintained by the resource is available to the user. Hence, the OS protects the electronic device and/or resources from unauthorized accesses, and, thereby, from unauthorized operations (read, write/modify, etc.) not intended by the user.

The endpoint or network account control server may maintain access (privilege) tokens associated with its users. Threat actors use a variation of token stealing to increase system privileges (e.g., from lower levels of privilege to administrator levels), whereby the threat actor may be provided access to privileged resources (e.g., protected files). By way of example of privilege level, Intel® X86 architectures running Windows® operating systems have typically four levels of privilege, with Level #0 being the highest privilege level with the broadest access rights and Level #3 being the lowest with narrow access rights. An electronic device operator may have a low privilege level, while a system administrator responsible for managing hundreds of such electronic devices, including that of the operator, may have a high privilege level. Examples of various privileges include permission to CREATE a file in a directory, or to READ or DELETE a file, access an external device, or to READ or WRITE to a socket for communicating over the Internet.

Previous systems for determining token stealing, such as pre-configured hypervisors for example, are resource intensive. Hence, hypervisors may cause user operated endpoints (or other electronic devices with constrained resources) to experience unacceptable data processing latencies, which adversely effects the user's overall experience in using the endpoint (or electronic device).

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the disclosure are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:

FIG. 1 is an exemplary block diagram of a general, physical representation of logic components of a token analysis system illustrated within an electronic device.

FIG. 2 is an exemplary embodiment of a logical representation of the logic components of the token analysis system of FIG. 1.

FIG. 3A is an exemplary embodiment of the access token data store accessible by the kernel mode agent of FIG. 2 and providing storage for the stored token state information.

FIG. 3B is an exemplary embodiment of current token state information associated with a current copy of the access token being part of an event for subsequent comparison to the stored token state information of FIG. 3A.

FIG. 4 is an exemplary embodiment of a flowchart illustrating the operations of detecting and analyzing a privilege escalation detected by the kernel mode driver of FIG. 2 and a notification of the detection being provided to the user mode agent of FIG. 2.

FIG. 5 is an exemplary embodiment of a flowchart illustrating operations conducted by the kernel mode driver of FIG. 2 in detecting a privilege escalation.

 DETAILED DESCRIPTION I. Overview Summary

Embodiments of the disclosure relate to a token analysis system and method for detecting privilege escalation within an access token and determining whether the privilege escalation is part of a cyberattack. This particular type of cyberattack is referred to as a “privilege escalation attack.” According to one embodiment of the disclosure, the computerized method starts with the storing of metadata for an access token in response to detecting a launch of a process being monitored, where the metadata includes token state information. Herein, for this embodiment, the token state information may include (i) an identifier of the user to which the access token pertains (e.g., user identifier), and (ii) a pointer to the access token data structure including a set of privileges provided to the user from the access token. This stored token state information is then available as a baseline for comparison purposes, as described below. Hence, the stored token state information, being a stored copy of contents of the access token, may also be referred to as a “baseline token snapshot.”

It is noted that, for some operating environments, the token analysis system may maintain separate copies of baseline token snapshots (stored token state information) on aper-process basis. Alternatively, the token analysis system may maintain a single copy of the baseline token snapshot on a per token basis, which may be referenced by pointers utilized by processes in other operating environments. Stated differently, multiple baseline token snapshots corresponding to different processes may be used, or alternatively, a single baseline token snapshot is used, but pointers to the same (for all processes) baseline token snapshot may be used.

More specifically, as described below, different token storage schemes may be deployed to ensure that the baseline token snapshot (i.e., stored token state information) is trusted. As a first token storage scheme, a process runs in the context of a user, and thus, the process is assigned an access privilege token at launch. Additionally, during launch, a baseline token snapshot is generated. At this processing stage, the baseline token snapshot is considered to be “trusted” because the access token is not yet being used in accessing resources. The content within the current access token (hereinafter, “current token state information”) may be analyzed periodically or aperiodically during execution of the process to ensure that no unauthorized changes to the access token have been made. Such analysis may include retrieval of at least a portion of the current token state information for comparison with at least a portion of the stored token state information corresponding to the baseline token snapshot.

In contrast, as a second token storage scheme, the baseline token snapshot is generated prior to launch of the process. Herein, the baseline token snapshot may be generated during or in response to a condition in which the user gains control to an electronic device. For instance, as an illustrative example, the baseline token snapshot may be generated during or in response to a log-on event by the user. As creation of the baseline token snapshot occurs prior to launch, and again before usage of the access token in gaining access to access-controlled resources, the baseline token snapshot is considered “trusted”.

For both the first and second token storage schemes, subsequent (substitute or revised) baseline token snapshots may be generated periodically or upon the occurrence of certain captured events, e.g., when a legitimate request for change to access privileges is requested or granted. For example, access token privileges may be analyzed periodically or aperiodically during execution of a process. The analysis includes the retrieval of the content associated with the current access token (e.g., access privileges, user identifier, etc.) for comparison with corresponding content from the baseline token snapshot (also referred to as the “stored token state information”).

Operating system functions (or other function calls) may be executed to change the access token legitimately, and if the access token is being changed using a legitimate function, the baseline token snapshot is revised to produce a new baseline token snapshot for later comparison with access token content. Otherwise, the method continues by detecting a privilege escalation attack, if underway, in response to a variation between the current token state information and the stored token state information and no legitimate functions are not called, which denotes a potential privilege escalation attack. The retrieval and comparison of the content of the current access token with the stored token state information (baseline token snapshot) may occur, in some embodiments for example, at the time of a file access event or other event.

Upon detecting a privilege escalation attack, the token analysis system may be adapted to terminate the malicious process and/or initiate an alert to an endpoint operator and/or, etc.) to nan administrator of a detected escalation attack (e.g., displayed warning, text message, email message, or automated phone call). Optionally, the event analysis logic within the token analysis system may access the access token data store maintained within the kernel space to reset one or more privileges associated with the compromised access token to its intended value, as stored prior to the attack.

More specifically, and according to one embodiment of the disclosure, the token analysis system includes software running in the user space and software running in the kernel space within memory of an electronic device. The “kernel space” is a portion of memory reserved for running a privileged operating system (OS) kernel, kernel extensions, and most device drivers. In contrast, “user space” is a portion of memory where application software executes outside of the kernel space. The token analysis system includes a first component operating in the user space and a second component operating in the kernel space of the electronic device.

Herein, according to one embodiment of the disclosure, the first component includes analysis logic that operates in accordance with a plurality of rules to determine whether a detected privilege escalation is malicious, suspicious or benign. The first component further provides a listing of processes to be monitored. Stored in the user space, the monitored process listing may be modifiable, where listed processes may be added, changed or removed by a security administrator via an input/output (I/O) interface or automatically by software running on the first component updating the listed processes based on white list and black list processes as described below. The addition, change and/or removal of a monitored process may depend, at least in part, on the current threat landscape to which the electronic device is exposed. The current threat landscape may be learned from threat intelligence, including prior privilege escalation attacks detected by the electronic device or by external sources such as (i) another electronic device communicatively coupled to the electronic device or (ii) a management system monitoring operations of the electronic device and/or other electronic devices within an enterprise. Additionally, or in the alternative, the threat intelligence may further include information gathered by a third party source (referred to as “third-party data”).

The second component includes event analysis logic, which is configured to (i) monitor for certain processes to obtain the initial access token associated with each process, (ii) monitor for certain events associated with the monitored processes identified by the monitored process listing, and (iii) detect changes to access tokens that modify the level of access to resources. In general, an “event” refers to an operation, task or activity that is performed by a process running on the electronic device. The selection of events to be monitored may be based on experiential knowledge and machine learning. In some situations, the event may be undesired or unexpected, indicating a potential cyberattack is being attempted. Examples of general events that are more susceptible to a cyberattack (and tend to be some of the monitored processes) may include, but are not limited or restricted to a file operation, registry operation, and/or thread creation, as described below.

At start-up for each monitored process (e.g., launch), metadata for an access token that specifies the user and access privileges conferred to that monitored process may be extracted and stored in memory accessible to the second component. The metadata may include and/or provide access to (i) a pointer to the access token data structure and/or (ii) contents of the access token (e.g., user identifier and/or access privileges) for later analysis. As described below, the metadata may be referred to as “stored token state information.” Thereafter, in response to detecting one or more selected events during operation of the monitored process, the current token state information (e.g., any token information suitable for use in determining a change in content advantageous in privilege escalation such as token pointer, user identifier, and/or any access privilege) is extracted by the second component and compared to corresponding content of the stored token state information. Privilege escalation is detected based on differences between the current token state information and the stored token state information. Given that not all privileged escalations are malicious, in some situations, the second component may resort to light-weight heuristics or another detection measure to determine whether a privilege escalation is part of a cyberattack, as described below.

Upon detection of privilege escalation and the privilege escalation is determined to be malicious, the first component may cause remedial action to be taken. The first component may cause the processor to terminate the monitored process (and perhaps all processes) to prevent any further malicious activities on the endpoint. Additionally, or in the alternative, the first component may generate an alert to the user or an administrator, reset the access token to its original privilege levels, and quarantine the object that caused the execution of the malicious process by placement of the object in an isolated segment of memory for subsequent analysis and deletion. As used herein, the term “quarantine” may refer to a temporary or permanent halting in processing of the object (e.g., a file, an application or other binary, a Uniform Resource Locator “URL,” etc.) initiated by the first component.

More specifically, the token analysis system monitors the operations associated with selected processes. In some embodiments, depending on the availability of endpoint resources, these monitored processes may be a limited subset of those processes running on the endpoint to avoid negatively affecting user experience on the endpoint. However, in some situations, the monitored processes may constitute all of the processes running on the electronic device. The selection of which processes to monitor may be based on a whitelist of typically benign processes and/or a blacklist of processes that are more commonly subverted to malicious activities found in the current threat landscape. The whitelist and blacklist constitute threat intelligence, which may be downloaded into the token analysis system or simply made available to (and accessible by) the token analysis system.

As described above, the token analysis system captures privilege escalations associated with the monitored processes and evaluates those changes to determine whether they are likely part of a privilege escalation attack. According to one embodiment of the disclosure, the first component may correspond to a software agent deployed within the user space of an electronic device (hereinafter, “user mode agent”), which is configured to determine whether a detected privilege escalation is part of a privilege escalation attack. Furthermore, the second component may correspond to a software driver deployed within the kernel space of the electronic device (hereinafter, “kernel mode driver”), which is configured to capture OS notifications directed to accesses of privileged resources and detect the privilege escalation.

For example, according to one embodiment of the disclosure, the kernel mode driver may receive one or more response messages (e.g., callback) directed to one or more requests (e.g., captured or hooked events) associated with monitored processes that are seeking access to a resource using privileges maintained within an access token. The user mode agent analyzes information associated with a privilege escalation detected by the kernel mode driver to determine a threat level of the privilege escalation (e.g., malicious, suspicious or benign). The user mode agent may be configured to issue an alert upon detecting a privileged escalation attack or request information associated with additional captured events produced by the monitored process involved in a suspicious privilege escalation in order to discern if the privilege escalation is malicious or benign.

To determine if a privilege escalation should be classified as malicious or not, the user mode agent may employ (run) heuristics (rules) with respect to information associated with the monitored process and/or event that attempted the privilege escalation. These privilege escalation rules may be tailored to the type of monitored process and further tailored to particular events associated with the monitored process (e.g., process create/terminate events, configuration change events, etc.). In some situations, the events may pertain to the monitored process and any “child” processes (at any tier) resulting therefrom. The heuristics may be used to validate events with monitored processes involved in the detected privilege escalation against a known set of benign and/or malicious events. If the heuristics indicate the events should be deemed malicious, the user mode agent may report a cyberattack is occurring and/or take remedial action. In some embodiments, the analysis of captured events may be conducted by a remote appliance, facility or cloud service rather than by the user mode agent on the endpoint.

It is noted that the determination whether the maliciousness of the monitored process can be based on privilege changes alone, or maliciousness may be based on privilege changes in conjunction with additional suspicious captured events, such as opening a socket for outbound communication after a privilege exception.

The monitored events may include software calls, such as Application Programming Interface (API) calls, system calls and the like. According to another embodiment, the user mode agent may capture (e.g., intercept or hook) any API calls issued by a monitored process, and the kernel mode driver may monitor OS notifications (in response to API calls) to the monitored process that may modify an access token, including privileges escalation. Hence, the user mode agent may determine whether a privilege escalation occurred based on events other than a legitimate API or software call, which may be indicative of maliciousness. In other words, while such APIs are often used to request, legitimately, a change in privilege, if monitoring of the token privileges identifies the change in privilege without use of such an API, the privilege escalation should be deemed at least suspicious of being for malicious purposes.

II. Terminology

In the following description, certain terminology is used to describe various features of the invention. For example, each of the terms “logic” and “component” may be representative of hardware, firmware or software that is configured to perform one or more functions. As hardware, the logic (or component) may include circuitry having data processing and/or storage functionality. Examples of such circuitry may include, but are not limited or restricted to a hardware processor (e.g., microprocessor, one or more processor cores, a digital signal processor, a programmable gate array, a microcontroller, an application specific integrated circuit “ASIC”, etc.), a semiconductor memory, or combinatorial elements.

Additionally, or in the alternative, the logic (or component) may include software such as one or more processes, one or more instances, Application Programming Interface(s) (API), subroutine(s), function(s), applet(s), servlet(s), routine(s), source code, object code, shared library/dynamic link library (dll), or even one or more instructions. This software may be stored in any type of a suitable non-transitory storage medium, or transitory storage medium (e.g., electrical, optical, acoustical or other form of propagated signals such as carrier waves, infrared signals, or digital signals). Examples of a non-transitory storage medium may include, but are not limited or restricted to a programmable circuit; non-persistent storage such as volatile memory (e.g., any type of random access memory “RAM”); or persistent storage such as non-volatile memory (e.g., read-only memory “ROM”, power-backed RAM, flash memory, phase-change memory, etc.), a solid-state drive, hard disk drive, an optical disc drive, or a portable memory device. As firmware, the logic (or component) may be stored in persistent storage.

Herein, a “communication” generally refers to related data that is received, transmitted, or exchanged within a communication session. This data may be received, transmitted, or exchanged in the form a message, namely information in a prescribed format and transmitted in accordance with a suitable delivery protocol. A “message” may be in the form of one or more packets, frames, or any other series of bytes or bits having the prescribed format.

The term “computerized” generally represents that any corresponding operations are conducted by hardware in combination with software and/or firmware.

The term “agent” generally refers to a module of software installed on a target system (e.g., an endpoint) that enables a user (e.g., a human such as an operator of an electronic device, an administrator or an external computer system) to monitor and interact with the target system. Agents allow users to gather information about multiple aspects of the target system. In some embodiments, agents also permit users to remotely retrieve the captured events and select other contents of the target system's memory or hard drive, and could potentially be configured to modify its security rules, configuration information and select other content. The agent may be configured to either communicate over a computer network, or to read and write all relevant configuration information and acquired data to a computer storage medium, such as a hard drive or removable read/write media (USB key, etc.). In one embodiment, the agent may be built in a modular fashion. The ability to gather a particular piece of data from a target system (e.g. a list of running processes on the target system) is implemented as a discrete module of software and loaded by the agent. This allows for easy adaptation of the agent to different environments that have specific requirements for data collection.

According to one embodiment of the disclosure, the term “malware” may be broadly construed as any code, communication or activity that initiates or furthers a cyber-attack. Malware may prompt or cause unauthorized, anomalous, unintended and/or unwanted behaviors or operations constituting a security compromise of information infrastructure. For instance, malware may correspond to a type of malicious computer code that, as an illustrative example, executes an exploit to take advantage of a vulnerability in a network, electronic device or software, for example, to gain unauthorized access, harm or co-opt operation of an electronic device or misappropriate, modify or delete data. Alternatively, as another illustrative example, malware may correspond to information (e.g., executable code, script(s), data, command(s), etc.) that is designed to cause an electronic device to experience anomalous (unexpected or undesirable) behaviors. The anomalous behaviors may include a communication-based anomaly or an execution-based anomaly, which, for example, could (1) alter the functionality of an electronic device executing application software in an atypical manner; (2) alter the functionality of the electronic device executing that application software without any malicious intent; and/or (3) provide unwanted functionality which may be generally acceptable in another context.

The term “electronic device” may be construed as any computing system with the capability of processing data and/or connecting to a network. Such a network may be a public network such as the Internet or a private network such as a wireless data telecommunication network, wide area network, a type of local area network (LAN), or a combination of networks. Examples of an electronic device may include, but are not limited or restricted to, an endpoint (e.g., a laptop, a mobile phone, a tablet, a computer, a wearable such as a smartwatch, Google® Glass, health monitoring device, or the like), a standalone appliance, a server, a video game console, a set top box, a smart (networked) home appliance, a router or other intermediary communication device, a firewall, etc.

The term “transmission medium” may be construed as a physical or logical communication path between two or more electronic devices or between components within an electronic device. For instance, as a physical communication path, wired and/or wireless interconnects in the form of electrical wiring, optical fiber, cable, bus trace, or a wireless channel using radio frequency (RF) or infrared (IR), may be used. A logical communication path may simply represent a communication path between two or more electronic devices or between components within an electronic device.

The term “privilege level” refers to the delegated authority (permissions) of a user to cause (e.g., via one or more processes), an operation, task or an activity to be performed on an electronic device. A user obtains a grant of privileges by presenting credentials to a privilege-granting authority. This may be accomplished by the user logging onto a system with a username and password, and if the username and password supplied are approved, the user is granted privileges as described a certain level of privileges. Such operations or tasks are tagged with a privilege level required for them to be permitted to be performed (demanded privilege). When a task tries to access a resource, or execute a privileged instruction, the processor determines whether the user making the request has the demanded privilege and, if so, access is permitted; otherwise, a “protection fault” interrupt is typically generated. Accordingly, for malware to succeed in gaining access to protected (privileged) resources or otherwise executing privileged instructions, the malware often requires escalated privilege.

Finally, the terms “or” and “and/or” as used herein are to be interpreted as inclusive or meaning any one or any combination. Therefore, “A, B or C” or “A, B and/or C” mean “any of the following: A; B; C; A and B; A and C; B and C; A, B and C.” An exception to this definition will occur only when a combination of elements, functions, steps or acts are in some way inherently mutually exclusive.

As this invention is susceptible to embodiments of many different forms, it is intended that the present disclosure is to be considered as an example of the principles of the invention and not intended to limit the invention to the specific embodiments shown and described.

III. General Architecture

Referring now to FIG. 1, an exemplary block diagram of a physical representation of components of a token analysis system 110 deployed within an electronic device 100 is shown. Herein, the token analysis system 110 may be installed within the electronic device 100, which may correspond to a physical endpoint or a virtual endpoint. A “virtual endpoint” includes a run-time environment operating within a virtual machine (hosted on an electronic device) that simulates operations of a physical endpoint, where one or more events produced during runtime of an object are monitored and analyzed as to whether such events are part of a cyberattack (e.g., a privilege escalation attack). A “physical” endpoint is an electronic device with electronics such as a laptop computer, a wearable, or the like.

As shown in FIG. 1, the token analysis system 110 is deployed within the endpoint 100. The endpoint 100 includes one or more processors 120 and persistent memory 130 (e.g., non-transitory computer-readable medium storage). The persistent memory 130 may include a user mode agent 140 and a kernel mode driver 150. The persistent memory 130 may further include one or more applications 1601-160M (M≥1) and an operating system (OS) 170. Herein, a processor 120 of the endpoint device 100 running the OS 170 (e.g., Windows® OS) has two different operating modes: a first operating mode and a second operating mode. The processor 120 is configured to switch between these two operating modes depending on what type of code is running on the processor 120. For instance, the user mode agent 140 and applications 1601-160M run during the first operating mode while the kernel mode driver 150 and certain OS components 170 run during second operating mode. The user mode agent 140 and the kernel mode driver 150 collectively operate to detect a modification of an access token (e.g., a privilege escalation) and determine whether the privilege escalation is part of a cyberattack.

When a user-mode application (e.g., application 1601) is launched, the processor 120 creates a process 180 (instance) for that application 1601. During run-time, the process 180 performs certain events. As described above, an “event” generally refers to a task or an activity that is performed by a process initiated by a software component, such as application 1601 for example, running on the endpoint (virtual or real). In some situations, the event may be undesired or unexpected, indicating a potential cyberattack is being attempted. Examples of such events may include, but are not limited or restricted to a file operation (e.g., open file, close file, etc.), registry operation (e.g., change registry key, change registry value, etc.), thread creation, or the like.

For certain processes being monitored by the kernel mode driver 150, content associated with events produced by the monitored processes, namely current state information associated with an access token (i.e., current token state information), may be evaluated (e.g., compared) to previously stored (cached) state information associated with the access token (i.e. stored token state information) to determine whether the access token has been impermissibly altered. As described above, the stored token state information is then available as a baseline for comparison purposes, gathered during launching of the process (as described below) or during a pre-launch state (e.g., response to a log-on event, etc.). The current token state information may include an address pointer to a structure associated with the access token, a user identifier identifying the user associated with the access token, and/or access privilege parameters associated with the user.

The endpoint 100 includes one or more interfaces 180, which may include a network interface 185 and/or input/output (I/O) interfaces 190. According to this embodiment of the disclosure, these components are connected by a transmission medium 195, such as any type of interconnect (e.g., bus), are at least partially encased in a housing made entirely or partially of a rigid material (e.g., hardened plastic, metal, glass, composite, or any combination thereof). The housing protects these components from environmental conditions.

Referring to FIG. 2, an exemplary block diagram of operations of the token analysis system 110 is shown. The token analysis system 110 is shown located in both the user space 200 (e.g., upper section) and the kernel space 205 (e.g., lower section), e.g., includes components located in each space. The user mode agent 140 of the token analysis system 110 is located in user space 200 while the kernel mode driver 150 is located in kernel space 205. The exemplary interactions between the user mode agent 140 and the kernel mode driver 150 are referenced by circled numbers, which correspond to operations of one or more of the components illustrated.

Initially, according to one embodiment of the disclosure, the token analysis system 110 may be configured to monitor selected processes as to whether such processes are involved in a privilege escalation attack. As shown, the user mode agent 140 includes analysis logic 210, which is communicatively coupled to a data store 220 that maintains a monitored process listing 225. Herein, the data store 220 may be implemented as part of the user mode agent 140 or may be external from the user mode agent 140 and, for example, within or remote to (depending on the embodiment) the endpoint 100. According to this embodiment, the monitored process listing 225 identifies one or more processes that are selected to be monitored by the kernel mode driver 150 for privilege escalation. Herein, the processes within the monitored process listing 225 may be identified by specific process name, although other metadata may be used in identifying such processes (e.g., pointer to data structure of the process, etc.).

The monitored process listing 225 may be pre-loaded at installation of the token analysis system 110, and thereafter, the monitored processes within the listing 225 may be altered (e.g., added, removed, updated, etc.) from time to time. Hence, the data store 220 is communicatively coupled to receive information for updating the monitored process listing 225 via a network interface 185 or an I/O interface 190 as shown in FIG. 1 (operation 1). Such communicative coupling allows for monitored processes to be changed based on the threat landscape, which may be at least partially represented by threat analysis data received from another electronic device that is part of the same enterprise network (e.g., a management system gathering cybersecurity information from other electronic devices, etc.) or from a third party source over a public network (e.g., internet download, etc.).

More specifically, as shown in FIG. 2, the user mode agent 140 interacts with the kernel mode driver 150 to download the monitored process listing 225 on a synchronous or asynchronous basis (operation 2). For instance, the user mode agent 140 may download the monitored process listing 225 in response to an event, such as a notification of availability of an update (e.g., removal, change, or addition of a monitored process to the listing 225). Additionally, or in the alternative, the user mode agent 140 may download the monitored process listing 225 on a periodic (synchronous) basis (e.g., at a prescribed time, after a prescribed time period has elapsed from the last download, etc.). The kernel mode driver 150 stores the monitored process listing 225 in a data store 230 accessible by event analysis logic 250 of the kernel mode driver 150.

Thereafter, the kernel mode driver 150 subscribes to the Operating System (OS) of the electronic device 100 to receive information associated with events for each of the monitored processes of the process listing 225 (operation 3). In particular, the kernel mode driver 150 initiates a request message 235 via an Application Programming Interfaces (API) to the OS of the electronic device 100, which sets the OS to generate a response message (e.g., callback) to the kernel mode driver 150 in response to a monitored process being launched (operation 4) and certain subsequent events being performed while the monitored process is active (operations 8, 12 or 16). An example of one of the APIs may include PsSetCreateProcessNotifyRoutine. As a result, response messages (callbacks) for events performed by one or more monitored processes may be further received in response to selected events being attempted by the monitored process, such as a file operation, a registry operation, a thread creation, or another selected event being conducted during the monitored process.

For instance, as shown in FIG. 2, in response to a monitored process being launched under a particular user context, the kernel mode driver 150 receives a response message (e.g., callback) 240 associated with a creation of the process (operation 5). The event analysis logic 250 within the kernel mode driver 150 accesses the monitored process listing 225 stored in the data store 230 to confirm that the process identified in the callback message 240 is one of the monitored processes (operation 6). Upon determining that the process is one of the monitored processes, the event analysis logic 250 of the kernel mode driver 150 effectively stores a key value pair 260 associated with the monitored process into an access token data store 265 (operation 7). Each key value pair (e.g., key value pair 260) includes (or provides access to) stored token state information (CTSI) 320 that features the initial privileges maintained by the access token, which identifies the security context (demanded privilege) of the process.

More specifically, the stored token state information 320 operates as a “baseline token snapshot” to capture a trusted copy of contents of the access token. For one embodiment, during launch of the monitored process the stored token state information (baseline token snapshot) 320 is generated. The stored token state information 320 is considered to be “trusted” because the access token has not yet being used in accessing resources. It is noted that, for some operating environments, the token analysis system 110 may maintain separate copies of stored token state information 320 on a per-process basis. Alternatively, the token analysis system 110 may maintain a single copy of the stored token state information 320 on a per token basis, which may be referenced by pointers utilized by processes in other operating environments. However, for another embodiment, in lieu of capturing the stored token state information during launching of the monitored process, the stored token state information 320 may be generated before such launching. For example, the stored token state information 320 (baseline token snapshot) may be generated during or in response to a condition in which the user gains control to an electronic device, such as during a user log-on process.

Independent as to when the “trusted” baseline token snapshot 320 is captured, subsequent (substitute or revised) baseline token snapshots may be generated periodically or upon the occurrence of certain captured events. One type of event may be in response to a legitimate request for change to access privileges is requested or granted. Hence, access token privileges may be analyzed periodically or aperiodically during execution of a process, and the stored token state information 320 may be updated accordingly.

Referring now to FIG. 3A, an exemplary embodiment of storage of the stored token state information within the access token data store 265 is shown. The key value pair 260 represents the initial token state of the monitored process, including a pointer 300 to a data structure 310 for the monitored process as allocated by the OS (e.g., the “EPROCESS” data structure for Windows® OS) and one or more parameters 320 associated with an access token for the monitored process, namely the stored token state information that is accessible from the monitored process data structure 310. The stored token state information 320 may include, but are not limited or restricted to, a pointer 330 to an access token data structure 340, an identifier of the user (User ID 350) associated with the access token maintained within the access token data structure 340, and privileges 360 maintained within the access token data structure 340 that apply to the user identified by the User ID 350. The User ID 350 and privileges 360 collectively identify the level of access afforded to the user of endpoint and/or network resources. It is contemplated that the pointer 330, User ID 350 and privileges 360 may be stored together or in accordance with any storage nesting scheme.

Referring back to FIG. 2, in response to a selected event performed by the process being monitored, such as a file operation (e.g., open file) for example, a file operation callback message 270 is provided to the kernel mode driver 150 (operations 8&9). The file operation callback message 270 includes the pointer 300 to the monitored process data structure 310, from which the event analysis logic 250 may obtain the current token state information 272 for the monitored process for comparison to the stored token state information 320 retrieved from the access token data store 265 (operations 10&11). Based on this comparison, the event analysis logic 250 is able to detect a potential privilege escalation attack.

As an illustrative example, as shown in FIG. 3B, from the file operation callback message 270, the event analysis logic 250 may obtain the current token state information 272 associated with a current copy of the access token 265 relied upon for the file operation. The event analysis logic 250 may further obtain a current pointer 274 to the access token data structure accessible via content within the file operation callback message 270; a user identifier 276 accessible via content within the file operation callback message 270; and current privileges 278 of the current copy of the access token 265 assigned accessible within the access token data structure and assigned to the user identified by the user identifier 276. Thereafter, the event analysis logic 250 compares (i) the current pointer 274 to the stored pointer 300 of FIG. 3A, (ii) the user identifier 276 to the User ID 350 of FIG. 3A, and (iii) the current privileges 278 to the stored privileges 360 of FIG. 3A.

More specifically, in response to the content of the current pointer 274 being altered and different from the content of the pointer 300 of FIG. 3A stored at commencement of the monitored process or the content of the user identifier 276 being altered and different from the content of the User ID 350 of FIG. 3A, the event analysis logic 250 transmits a detection message to the user mode agent 140 of the potential privilege escalation attack. Additionally, in response to the content of the current privileges 278 being different than the stored privileges 360 of FIG. 3A, the event analysis logic 250 transmits a detection message to the user mode agent 140 of the potential privilege escalation attack as illustrated in FIG. 4.

Referring back to FIG. 2, in response to another selected event performed by the monitored process, such as a registry modification (operation 12) or a thread creation (operation 16) for example, a callback message 280 for the registry modification (operation 13) or a callback message 285 for the thread creation (operation 17) is provided to the kernel mode driver 150. Each of these callback messages 280/290 may include a current pointer 282/292 to the monitored process data structure to read current token state information for the monitored process (e.g., pointer 284/294 to the access token data structure; user identifier 286/296, privileges 288/298), which is compared against the stored token state information 320 (e.g., stored pointer 300, User ID 350 and/or stored privileges 360 of FIG. 3A), which are retrieved from the access token data store 265 (operation 14 or operation 18). As before, based on uncovered changes in the current token state information, such as access token pointer value, the user identifier or stored privileges for the identified user for example, the event analysis logic 250 is able to detect a potential privilege escalation attack (operation 15 or operation 19).

Referring now to FIG. 4, an exemplary embodiment of the operations conducted for detecting a privilege escalation by the kernel mode driver 150 and analysis of the privilege escalation by the user mode agent 140 of FIG. 2 is shown. Upon detection of a potential privilege escalation, the kernel mode driver 150 uploads a detection message 400 to the user mode agent 140 (operation 20). The detection message 400 represents detection of a potential privilege escalation attack requiring further analysis. The detection message 400 features a data structure including (i) information 410 identifies the monitored process (e.g., process name or process identifier), (ii) information 420 that identifies the event in which a change to one or more parameters of the access token was detected (e.g., event name or event identifier), and (iii) information 430 that identifies changes to token state information (e.g., metadata associated with the access token). The token state information 430 may include information associated with (a) changes made from the access token pointer 300 for accessing the access token data structure 340, (b) changes made from the User ID 350, and/or (c) changes made from the stored privileges 360. Additionally, it is contemplated that the detection message 400 may further identify (d) the particular privileges 440 of the access token that are enabled and/or (e) the current User ID 450, where the user identifier has not changed.

Upon receipt of the detection message 400, the analysis logic 210 of the user mode agent 140 extracts the information 410, 420 and 430 and determines a threat level of the detected privilege escalation in accordance with the privilege escalation rules 460 that control operability of the analysis logic 210 (operation 21). According to one embodiment of the disclosure, the analysis logic 210 applies portions of the extracted information 410, 420 and/or 430 to the privilege escalation rules 460 to determine a threat level for the monitored process (operation 20). The threat level may be categorized as (1) “benign” (e.g., a confirmed legitimate privilege escalation); (2) “malicious” (e.g., a confirmed unauthorized privilege escalation associated with a privilege escalation attack); or (3) “suspicious” (e.g., an unauthorized privilege escalation but indeterminate of malicious intent).

Referring still to FIG. 4, responsive to a privilege escalation detection caused by a change in the access token data structure pointer and/or a change in the user identifier, the analysis logic 210 concludes that the monitored process (and monitored event associated the monitored process) is associated with a privilege escalation attack. However, the analysis logic 210 may conclude a detected privilege escalation is suspicious or not malicious, depending on the type of privilege(s) being modified and/or scope of such modification. As an illustrative example, a minor privilege change (e.g., guest-to-employee; one super-privilege level to another super-privilege level) may warrant a “suspicious” classification for increased surveillance, but does not rise to a “malicious” classification as the enhanced privileges fail to provide additional access to sensitive information within the endpoint or stored within the enterprise. Similarly, a major privilege change (e.g., employee-to-administrator; employee-to-super-privilege) may warrant a “malicious” classification as the enhanced privileges may provide the user with substantial increased access sensitive information within the endpoint or stored within the enterprise.

Once the analysis logic 210, operating in accordance with the privilege escalation rules 460, determines that a detected privilege escalation represented by the detection message 400 is malicious, the user mode agent 140 may initiate an alert 470 to the endpoint user or an administrator as to the detection of a privilege escalation attack (operation 22). Furthermore, besides the alert 470, the user mode agent 140 may terminate and/or quarantine the malicious monitored process.

In contrast, where the analysis logic 210 determines that the detected privilege escalation is suspicious, the user mode agent 140 may initiate an event acquisition message 480 to the kernel mode driver 150 to acquire additional events associated with the monitored process for evaluation (operation 23). As the kernel mode agent 150 publishes the subsequent events (such as, for example, access using the escalated privilege to a privileged and highly sensitive file or attempted outbound transfer of data (exfiltration) from that file) to the user mode agent 140, the user mode agent 140, operating in accordance with the privilege escalation rules, may identify further suspicious or malicious activity. The suspicious determination (described above) along with these additional results may be weighted, and collectively, may prompt the analysis logic 210 of the user mode agent 140 to determine a malicious verdict or a benign verdict.

Referring to FIG. 5, an exemplary embodiment of a flowchart illustrating operations of the token analysis system 110 of FIG. 2 in analysis of a process creation event for a monitored process is shown. Responsive to receiving a callback message directed to creation of a monitored process (block 500), the kernel mode driver determines or confirms that the process is a process selected for monitoring for token privilege escalation (operation 510). Where the process is not to be monitored, the kernel mode driver ignores the process (operation 515). However, where the process is a process to be monitored, the kernel mode driver provides the content associated with the monitored process to the event analysis logic 250 that performs the token privilege escalation detection. From the content, the kernel mode driver attempts to verify whether the monitored process is a “parent” process (operation 520). Stated differently, the kernel mode driver determines whether the monitored process is created from another process (i.e., a “child” process) or not.

In the event that the monitored process is a first process (i.e., a “parent” process) the kernel mode driver analyzes the current token state information associated with the access token for the created, monitored parent process (operations 525 and 550). However, where the monitored process is a “child” process, namely the monitored process is a secondary process created from another process, the kernel mode driver retrieves and analyzes the current token state information associated with the access token for the parent process (operations 530 and 540), which, by inherence, is also associated with the child process. If a privilege escalation is detected for the parent process, the detected privilege escalation of the child process is reported to the user mode agent (operation 545). However, if no privilege escalation is detected for the parent process, the kernel mode driver analyzes the current token state information associated with the access token for the created (child) process (operations 540 and 550), since the access token for the child process may have been changed by a user after created. If a privilege escalation is detected for the child process, the detected privilege escalation is reported to the user mode agent (operations 545 and 560). However, if no privilege escalation is detected for the monitored process, the kernel mode driver refrains from reporting information associated with the monitored process to the user model agent (operations 560 and 570).

In the foregoing description, the invention is described with reference to specific exemplary embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims.

Claims

1. A computerized method for detecting and mitigating a privilege escalation attack on an electronic device, the method comprising:

storing metadata for an access token, by a kernel mode driver, in response to an event occurring before or during a launch of a process being monitored, the stored metadata comprises stored token state information including content to provide access to at least a first set of privileges for a user associated with the access token;
responsive to detecting an event associated with the process being monitored, extracting at least a portion of current token state information for the access token by the kernel mode driver and comparing at least the portion of the current token state information to at least a portion of the stored token state information;
determining whether the access token has been changed legitimately by a software function that, upon execution, is permitted to change the access token;
detecting a privilege escalation attack in response to a variation between at least the portion of the current token state information and at least the portion of the stored token state information when the access token has not been changed legitimately, wherein the detecting of the privilege escalation attack being conducted by a user mode agent in response to determining a prescribed threat level of a detected privilege escalation based on (i) a difference between a pointer to a data structure of the access token included in the current token state information and a pointer to the data structure of the access token included in the stored token state information, or (ii) a difference between a user identifier included in the current token state information and a user identifier included in the stored token state information, or (iii) a difference between a set of privileges associated with the access token obtained via the pointer to the data structure of the access token included in the current token state information and the first set of privileges; and
revising the stored token state information when the access token has been changed legitimately.

2. The computerized method of claim 1, wherein the storing of the stored token state information is in response to the launch of the process being monitored.

3. The computerized method of claim 2, wherein prior to detecting the launch of the process, the method further comprising:

receiving a listing of monitored processes by the kernel mode driver.

4. The computerized method of claim 3, wherein the listing of monitored processes is provided from the user mode agent to the kernel mode driver, the listing of monitored processes being modifiable depending on a current threat landscape to which the electronic device is exposed.

5. The computerized method of claim 3, wherein prior to detecting the launch activation of the process and after receiving the listing of monitored processes, the method further comprising:

subscribing, by the kernel mode driver, to an operating system of the electronic device to receive information associated with events for one or more processes within the listing of the monitored processes.

6. The computerized method of claim 5, wherein prior to detecting the launch of the process and after receiving the listing of monitored processes, the method further comprising:

issuing, by the kernel mode driver, a request message via an Application Programming Interface (API) to an operating system of the electronic device to receive information associated with events for a first process within the listing of the monitored processes; and
receiving a response message from the operating system in response to the first process being launched.

7. The computerized method of claim 6, wherein the request message includes the stored token state information providing access to the first set of privileges for the user.

8. The computerized method of claim 6, wherein the response message includes the stored token state information that comprises a pointer to a data structure associated with the access token including the first set of privileges and an identifier of the user associated with the access token.

9. The computerized method of claim 8, wherein the detecting of the privilege escalation attack is based, at least in part, on detecting the pointer to the data structure associated with the access token being different from a pointer to a data structure for the access token provided by the extracted portion of the current token state information.

10. The computerized method of claim 8, wherein the detecting of the privilege escalation attack is based, at least in part, on detecting the identifier of the user included in the data structure associated with the access token being different from a user identifier within the data structure for the access token provided by the extracted portion of the current token state information.

11. The computerized method of claim 7, wherein the detecting of the privilege escalation attack is based on detected changes between (i) parameter values of the access token obtained from the portion of current token state information for the access token and (ii) corresponding parameter values from the first set of privileges for the user associated with the access token.

12. The computerized method of claim 1, wherein the user mode agent corresponds to a software agent deployed within a user space of the electronic device.

13. The computerized method of claim 1, wherein the detecting of the privilege escalation attack being conducted by the user mode agent in response to analysis of information within a message provided to the user mode agent by the kernel mode driver, the message includes (i) information that identifies the process being monitored, (ii) information that identifies the event, and (iii) information that identifies a change between the portion of the stored token state information and the portion of the current token state information.

14. The computerized method of claim 1, wherein the information that identifies the change between the portion of the stored token state information and the portion of the current token state information comprises (i) any difference between the pointer to the data structure of the access token included in the current token state information and the pointer to the data structure of the access token included in the stored token state information, (ii) any difference between the user identifier included in the current token state information and the user identifier included in the stored token state information, and (iii) any difference between the set of privileges associated with the access token obtained via the pointer to the data structure of the access token included in the current token state information and the first set of privileges.

15. The electronic device of claim 14, wherein the kernel mode driver, when executed by the processor, to detect the privilege escalation attack being based, at least in part, on detecting the pointer to the data structure associated with the access token being different than the pointer to the data structure for the access token provided by the extracted portion of the current token state information.

16. The electronic device of claim 14, wherein the kernel mode driver, when executed by the processor, to detect the privilege escalation attack is based, at least in part, on detecting the identifier of the user included in the data structure associated with the access token being different from a user identifier within a data structure for the access token provided by the extracted portion of the current token state information.

17. The electronic device of claim 14, wherein the user mode agent to detect the privilege escalation attack by at least detecting changes between (i) values of the access token obtained from the extracted portion of the current token state information and (ii) corresponding values from the stored token state information.

18. The electronic device of claim 14, wherein the user mode agent to detect the privilege escalation attack in response to at least determining a prescribed threat level of a detected privilege escalation based on (i) a difference between the pointer to a data structure of the access token included in the current state information and the pointer to the data structure of the access token included in the stored token state information, or (ii) a difference between the user identifier included in the current token state information and the user identifier included in the stored token state information.

19. The electronic device of claim 14, wherein the information that identifies the variation between at least the portion of the current token state information and at least the portion of the stored token state information comprises (i) any difference between the pointer to the data structure of the access token included in the current token state information and the pointer to the data structure of the access token included in the stored token state information, (ii) any difference between the user identifier included in the current token state information and the user identifier included in the stored token state information, and (iii) any difference between the set of privileges associated with the access token obtained via the pointer to the data structure of the access token included in the current token state information and the first set of privileges.

20. An electronic device, comprising:

a processor; and
a memory communicatively coupled to the processor, the memory includes a user mode agent and a kernel mode driver, wherein the kernel mode driver, when executed by the processor, to (i) store metadata for an access token prior to or during a launch of a process being monitored, the stored metadata comprises stored token state information including content to provide access to at least a first set of privileges for a user associated with the access token, (ii) extract at least a portion of current token state information for the access token in response to detecting an event associated with the process being monitored, (iii) compare at least the portion of the current token state information to at least a portion of the stored token state information, (iv) upon the compare identifies a change to at least the portion of the current token state information, determine whether the access token has been changed legitimately by a software function that, upon execution, is permitted to change the access token and revise the stored token state information for the access token if the access token has been changed legitimately, and the user mode agent, when executed by the processor, to detect a privilege escalation attack in response to a variation between at least the portion of the current token state information and at least the portion of the stored token state information, wherein the variation includes at least (i) a difference between a pointer to a data structure of the access token included in the current token state information and a pointer to the data structure of the access token included in the stored token state information, or (ii) a difference between a user identifier included in the current token state information and a user identifier included in the stored token state information, or (iii) a difference between a set of privileges associated with the access token obtained via the pointer to the data structure of the access token included in the current token state information and the first set of privileges.

21. The electronic device of claim 20, wherein the kernel mode driver being configured to receive a listing of monitored processes by the kernel mode driver.

22. The electronic device of claim 21, wherein user mode agent being configured to provide the listing of monitored processes to the kernel mode driver, the listing of monitored processes being modifiable depending on a current threat landscape to which the electronic device is exposed.

23. The electronic device of claim 21, wherein the kernel mode driver to subscribe to an operating system of the electronic device to receive information associated with events for one or more processes within the listing of the monitored processes.

24. The electronic device of claim 23, wherein the kernel mode driver to issue a request message via an Application Programming Interface (API) to an operating system of the electronic device to receive information associated with events for a first process within the listing of the monitored processes and receive a response message from the operating system in response to the first process being launched.

25. The electronic device of claim 20, wherein the request message issued by the kernel mode driver includes the stored token state information providing access to the first set of privileges for the user.

26. The electronic device of claim 25, wherein the response message received by the kernel mode driver includes the stored token state information that comprises the pointer to the data structure associated with the access token including the first set of privileges and an identifier of the user associated with the access token.

27. A non-transitory computer readable medium including software that, when executed by the processor, performs operations comprising:

storing token state information associated with an access token in response to an event occurring before or during a launch of a process being monitored, the stored token state information including content to provide access to at least a first set of privileges for a user associated with the access token;
responsive to detecting an event associated with the process being monitored, extracting at least a portion of current token state information for the access token and comparing at least the portion of the current token state information to at least a portion of the stored token state information;
determining whether the access token has been changed legitimately by a software function that, upon execution, is permitted to change the access token;
revising the stored token state information when the access token has been changed legitimately; and
detecting a privilege escalation attack in response to a variation between at least the portion of the current token state information and at least the portion of the stored token state information when the access token has not been changed legitimately, wherein the variation includes at least (i) a difference between a pointer to a data structure of the access token included in the current state information and a pointer to the data structure of the access token included in the token state information, or (ii) a difference between a user identifier included in the current state information and a user identifier included in the token state information, or (iii) a difference between a set of privileges associated with the access token obtained via the pointer to the data structure of the access token included in the current state information and the first set of privileges.
Referenced Cited
U.S. Patent Documents
4292580 September 29, 1981 Ott et al.
5175732 December 29, 1992 Hendel et al.
5319776 June 7, 1994 Hile et al.
5440723 August 8, 1995 Arnold et al.
5490249 February 6, 1996 Miller
5657473 August 12, 1997 Killean et al.
5802277 September 1, 1998 Cowlard
5842002 November 24, 1998 Schnurer et al.
5960170 September 28, 1999 Chen et al.
5978917 November 2, 1999 Chi
5983348 November 9, 1999 Ji
6088803 July 11, 2000 Tso et al.
6092194 July 18, 2000 Touboul
6094677 July 25, 2000 Capek et al.
6108799 August 22, 2000 Boulay et al.
6154844 November 28, 2000 Touboul et al.
6269330 July 31, 2001 Cidon et al.
6272641 August 7, 2001 Ji
6279113 August 21, 2001 Vaidya
6298445 October 2, 2001 Shostack et al.
6357008 March 12, 2002 Nachenberg
6424627 July 23, 2002 Sorhaug et al.
6442696 August 27, 2002 Wray et al.
6484315 November 19, 2002 Ziese
6487666 November 26, 2002 Shanklin et al.
6493756 December 10, 2002 O'Brien et al.
6550012 April 15, 2003 Villa et al.
6775657 August 10, 2004 Baker
6831893 December 14, 2004 Ben Nun et al.
6832367 December 14, 2004 Choi et al.
6895550 May 17, 2005 Kanchirayappa et al.
6898632 May 24, 2005 Gordy et al.
6907396 June 14, 2005 Muttik et al.
6941348 September 6, 2005 Petry
6971097 November 29, 2005 Wallman
6981279 December 27, 2005 Arnold et al.
7007107 February 28, 2006 Ivchenko et al.
7028179 April 11, 2006 Anderson et al.
7043757 May 9, 2006 Hoefelmeyer et al.
7058822 June 6, 2006 Edery et al.
7069316 June 27, 2006 Gryaznov
7080407 July 18, 2006 Zhao et al.
7080408 July 18, 2006 Pak et al.
7093002 August 15, 2006 Wolff et al.
7093239 August 15, 2006 van der Made
7096498 August 22, 2006 Judge
7100201 August 29, 2006 Izatt
7107617 September 12, 2006 Hursey et al.
7159149 January 2, 2007 Spiegel et al.
7213260 May 1, 2007 Judge
7231667 June 12, 2007 Jordan
7240364 July 3, 2007 Branscomb et al.
7240368 July 3, 2007 Roesch et al.
7243371 July 10, 2007 Kasper et al.
7249175 July 24, 2007 Donaldson
7287278 October 23, 2007 Liang
7308716 December 11, 2007 Danford et al.
7328453 February 5, 2008 Merkle, Jr. et al.
7346486 March 18, 2008 Ivancic et al.
7356736 April 8, 2008 Natvig
7386888 June 10, 2008 Liang et al.
7392542 June 24, 2008 Bucher
7418729 August 26, 2008 Szor
7428300 September 23, 2008 Drew et al.
7441272 October 21, 2008 Durham et al.
7448084 November 4, 2008 Apap et al.
7458098 November 25, 2008 Judge et al.
7464404 December 9, 2008 Carpenter et al.
7464407 December 9, 2008 Nakae et al.
7467408 December 16, 2008 O'Toole, Jr.
7478428 January 13, 2009 Thomlinson
7480773 January 20, 2009 Reed
7487543 February 3, 2009 Arnold et al.
7496960 February 24, 2009 Chen et al.
7496961 February 24, 2009 Zimmer et al.
7519990 April 14, 2009 Xie
7523493 April 21, 2009 Liang et al.
7530104 May 5, 2009 Thrower et al.
7540025 May 26, 2009 Tzadikario
7546638 June 9, 2009 Anderson et al.
7565550 July 21, 2009 Liang et al.
7568233 July 28, 2009 Szor et al.
7584455 September 1, 2009 Ball
7603715 October 13, 2009 Costa et al.
7607171 October 20, 2009 Marsden et al.
7639714 December 29, 2009 Stolfo et al.
7644441 January 5, 2010 Schmid et al.
7657419 February 2, 2010 van der Made
7676841 March 9, 2010 Sobchuk et al.
7698548 April 13, 2010 Shelest et al.
7707633 April 27, 2010 Danford et al.
7712136 May 4, 2010 Sprosts et al.
7730011 June 1, 2010 Deninger et al.
7739740 June 15, 2010 Nachenberg et al.
7779463 August 17, 2010 Stolfo et al.
7784097 August 24, 2010 Stolfo et al.
7832008 November 9, 2010 Kraemer
7836502 November 16, 2010 Zhao et al.
7849506 December 7, 2010 Dansey et al.
7854007 December 14, 2010 Sprosts et al.
7869073 January 11, 2011 Oshima
7877803 January 25, 2011 Enstone et al.
7904959 March 8, 2011 Sidiroglou et al.
7908660 March 15, 2011 Bahl
7930738 April 19, 2011 Petersen
7937387 May 3, 2011 Frazier et al.
7937761 May 3, 2011 Bennett
7949849 May 24, 2011 Lowe et al.
7996556 August 9, 2011 Raghavan et al.
7996836 August 9, 2011 McCorkendale et al.
7996904 August 9, 2011 Chiueh et al.
7996905 August 9, 2011 Arnold et al.
8006305 August 23, 2011 Aziz
8010667 August 30, 2011 Zhang et al.
8020206 September 13, 2011 Hubbard et al.
8028338 September 27, 2011 Schneider et al.
8042184 October 18, 2011 Batenin
8045094 October 25, 2011 Teragawa
8045458 October 25, 2011 Alperovitch et al.
8069484 November 29, 2011 McMillan et al.
8087086 December 27, 2011 Lai et al.
8171553 May 1, 2012 Aziz et al.
8176049 May 8, 2012 Deninger et al.
8176480 May 8, 2012 Spertus
8201246 June 12, 2012 Wu et al.
8204984 June 19, 2012 Aziz et al.
8214905 July 3, 2012 Doukhvalov et al.
8220055 July 10, 2012 Kennedy
8225288 July 17, 2012 Miller et al.
8225373 July 17, 2012 Kraemer
8233882 July 31, 2012 Rogel
8234640 July 31, 2012 Fitzgerald et al.
8234709 July 31, 2012 Viljoen et al.
8239944 August 7, 2012 Nachenberg et al.
8260914 September 4, 2012 Ranjan
8266091 September 11, 2012 Gubin et al.
8286251 October 9, 2012 Eker et al.
8291499 October 16, 2012 Aziz et al.
8307435 November 6, 2012 Mann et al.
8307443 November 6, 2012 Wang et al.
8312545 November 13, 2012 Fuvell et al.
8321936 November 27, 2012 Green et al.
8321941 November 27, 2012 Fuvell et al.
8332571 December 11, 2012 Edwards, Sr.
8365286 January 29, 2013 Poston
8365297 January 29, 2013 Parshin et al.
8370938 February 5, 2013 Daswani et al.
8370939 February 5, 2013 Zaitsev et al.
8375444 February 12, 2013 Aziz et al.
8381299 February 19, 2013 Stolfo et al.
8402529 March 19, 2013 Green et al.
8464340 June 11, 2013 Ahn et al.
8479174 July 2, 2013 Chiriac
8479276 July 2, 2013 Vaystikh et al.
8479291 July 2, 2013 Bodke
8510827 August 13, 2013 Leake et al.
8510828 August 13, 2013 Guo et al.
8510842 August 13, 2013 Amit et al.
8516478 August 20, 2013 Edwards et al.
8516590 August 20, 2013 Ranadive et al.
8516593 August 20, 2013 Aziz
8522348 August 27, 2013 Chen et al.
8528086 September 3, 2013 Aziz
8533824 September 10, 2013 Hutton et al.
8539582 September 17, 2013 Aziz et al.
8549638 October 1, 2013 Aziz
8555391 October 8, 2013 Demir et al.
8561177 October 15, 2013 Aziz et al.
8566476 October 22, 2013 Shiffer et al.
8566946 October 22, 2013 Aziz et al.
8584094 November 12, 2013 Dadhia et al.
8584234 November 12, 2013 Sobel et al.
8584239 November 12, 2013 Aziz et al.
8595834 November 26, 2013 Xie et al.
8627476 January 7, 2014 Satish et al.
8635696 January 21, 2014 Aziz
8682054 March 25, 2014 Xue et al.
8682812 March 25, 2014 Ranjan
8689333 April 1, 2014 Aziz
8695096 April 8, 2014 Zhang
8713631 April 29, 2014 Pavlyushchik
8713681 April 29, 2014 Silberman et al.
8726392 May 13, 2014 McCorkendale et al.
8739280 May 27, 2014 Chess et al.
8776229 July 8, 2014 Aziz
8782792 July 15, 2014 Bodke
8789172 July 22, 2014 Stolfo et al.
8789178 July 22, 2014 Kejriwal et al.
8793278 July 29, 2014 Frazier et al.
8793787 July 29, 2014 Ismael et al.
8805947 August 12, 2014 Kuzkin et al.
8806647 August 12, 2014 Daswani et al.
8832829 September 9, 2014 Manni et al.
8850570 September 30, 2014 Ramzan
8850571 September 30, 2014 Staniford et al.
8881234 November 4, 2014 Narasimhan et al.
8881271 November 4, 2014 Butler, II
8881282 November 4, 2014 Aziz et al.
8898788 November 25, 2014 Aziz et al.
8935779 January 13, 2015 Manni et al.
8949257 February 3, 2015 Shiffer et al.
8984638 March 17, 2015 Aziz et al.
8990939 March 24, 2015 Staniford et al.
8990944 March 24, 2015 Singh et al.
8997219 March 31, 2015 Staniford et al.
9009822 April 14, 2015 Ismael et al.
9009823 April 14, 2015 Ismael et al.
9027135 May 5, 2015 Aziz
9071638 June 30, 2015 Aziz et al.
9104867 August 11, 2015 Thioux et al.
9106630 August 11, 2015 Frazier et al.
9106694 August 11, 2015 Aziz et al.
9118715 August 25, 2015 Staniford et al.
9159035 October 13, 2015 Ismael et al.
9171160 October 27, 2015 Vincent et al.
9176843 November 3, 2015 Ismael et al.
9189627 November 17, 2015 Islam
9195829 November 24, 2015 Goradia et al.
9197664 November 24, 2015 Aziz et al.
9223972 December 29, 2015 Vincent et al.
9225740 December 29, 2015 Ismael et al.
9241010 January 19, 2016 Bennett et al.
9251343 February 2, 2016 Vincent et al.
9262635 February 16, 2016 Paithane et al.
9268936 February 23, 2016 Butler
9275229 March 1, 2016 LeMasters
9282109 March 8, 2016 Aziz et al.
9292686 March 22, 2016 Ismael et al.
9294501 March 22, 2016 Mesdaq et al.
9300686 March 29, 2016 Pidathala et al.
9306960 April 5, 2016 Aziz
9306974 April 5, 2016 Aziz et al.
9311479 April 12, 2016 Manni et al.
9355247 May 31, 2016 Thioux et al.
9356944 May 31, 2016 Aziz
9363280 June 7, 2016 Rivlin et al.
9367681 June 14, 2016 Ismael et al.
9398028 July 19, 2016 Karandikar et al.
9413781 August 9, 2016 Cunningham et al.
9426071 August 23, 2016 Caldejon et al.
9430646 August 30, 2016 Mushtaq et al.
9432389 August 30, 2016 Khalid et al.
9438613 September 6, 2016 Paithane et al.
9438622 September 6, 2016 Staniford et al.
9438623 September 6, 2016 Thioux et al.
9459901 October 4, 2016 Jung et al.
9467460 October 11, 2016 Otvagin et al.
9483644 November 1, 2016 Paithane et al.
9495180 November 15, 2016 Ismael
9497213 November 15, 2016 Thompson et al.
9507935 November 29, 2016 Ismael et al.
9516057 December 6, 2016 Aziz
9519782 December 13, 2016 Aziz et al.
9536091 January 3, 2017 Paithane et al.
9537972 January 3, 2017 Edwards et al.
9560059 January 31, 2017 Islam
9565202 February 7, 2017 Kindlund et al.
9591015 March 7, 2017 Amin et al.
9591020 March 7, 2017 Aziz
9594904 March 14, 2017 Jain et al.
9594905 March 14, 2017 Ismael et al.
9594912 March 14, 2017 Thioux et al.
9609007 March 28, 2017 Rivlin et al.
9626509 April 18, 2017 Khalid et al.
9628498 April 18, 2017 Aziz et al.
9628507 April 18, 2017 Haq et al.
9633134 April 25, 2017 Ross
9635039 April 25, 2017 Islam et al.
9641546 May 2, 2017 Manni et al.
9654485 May 16, 2017 Neumann
9661009 May 23, 2017 Karandikar et al.
9661018 May 23, 2017 Aziz
9674298 June 6, 2017 Edwards et al.
9680862 June 13, 2017 Ismael et al.
9690606 June 27, 2017 Ha et al.
9690933 June 27, 2017 Singh et al.
9690935 June 27, 2017 Shiffer et al.
9690936 June 27, 2017 Malik et al.
9736179 August 15, 2017 Ismael
9740857 August 22, 2017 Ismael et al.
9747446 August 29, 2017 Pidathala et al.
9756074 September 5, 2017 Aziz et al.
9773112 September 26, 2017 Rathor et al.
9781144 October 3, 2017 Otvagin et al.
9787700 October 10, 2017 Amin et al.
9787706 October 10, 2017 Otvagin et al.
9792196 October 17, 2017 Ismael et al.
9824209 November 21, 2017 Ismael et al.
9824211 November 21, 2017 Wilson
9824216 November 21, 2017 Khalid et al.
9825976 November 21, 2017 Gomez et al.
9825989 November 21, 2017 Mehra et al.
9838408 December 5, 2017 Karandikar et al.
9838411 December 5, 2017 Aziz
9838416 December 5, 2017 Aziz
9838417 December 5, 2017 Khalid et al.
9846776 December 19, 2017 Paithane et al.
9876701 January 23, 2018 Caldejon et al.
9888016 February 6, 2018 Amin et al.
9888019 February 6, 2018 Pidathala et al.
9910988 March 6, 2018 Vincent et al.
9912644 March 6, 2018 Cunningham
9912681 March 6, 2018 Ismael et al.
9912684 March 6, 2018 Aziz et al.
9912691 March 6, 2018 Mesdaq et al.
9912698 March 6, 2018 Thioux et al.
9916440 March 13, 2018 Paithane et al.
9921978 March 20, 2018 Chan et al.
9934376 April 3, 2018 Ismael
9934381 April 3, 2018 Kindlund et al.
9946568 April 17, 2018 Ismael et al.
9954890 April 24, 2018 Staniford et al.
9973531 May 15, 2018 Thioux
10002252 June 19, 2018 Ismael et al.
10019338 July 10, 2018 Goradia et al.
10019573 July 10, 2018 Silberman et al.
10025691 July 17, 2018 Ismael et al.
10025927 July 17, 2018 Khalid et al.
10027689 July 17, 2018 Rathor et al.
10027690 July 17, 2018 Aziz et al.
10027696 July 17, 2018 Rivlin et al.
10033747 July 24, 2018 Paithane et al.
10033748 July 24, 2018 Cunningham et al.
10033753 July 24, 2018 Islam et al.
10033759 July 24, 2018 Kabra et al.
10050998 August 14, 2018 Singh
10068091 September 4, 2018 Aziz et al.
10075455 September 11, 2018 Zafar et al.
10083302 September 25, 2018 Paithane et al.
10084813 September 25, 2018 Eyada
10089461 October 2, 2018 Ha et al.
10097573 October 9, 2018 Aziz
10104102 October 16, 2018 Neumann
10108446 October 23, 2018 Steinberg et al.
10121000 November 6, 2018 Rivlin et al.
10122746 November 6, 2018 Manni et al.
10133863 November 20, 2018 Bu et al.
10133866 November 20, 2018 Kumar et al.
10146810 December 4, 2018 Shiffer et al.
10148693 December 4, 2018 Singh et al.
10165000 December 25, 2018 Aziz et al.
10169585 January 1, 2019 Pilipenko et al.
10176321 January 8, 2019 Abbasi et al.
10181029 January 15, 2019 Ismael et al.
10191861 January 29, 2019 Steinberg et al.
10192052 January 29, 2019 Singh et al.
10198574 February 5, 2019 Thioux et al.
10200384 February 5, 2019 Mushtaq et al.
10210329 February 19, 2019 Malik et al.
10216927 February 26, 2019 Steinberg
10218740 February 26, 2019 Mesdaq et al.
10242185 March 26, 2019 Goradia
20010005889 June 28, 2001 Albrecht
20010047326 November 29, 2001 Broadbent et al.
20020018903 February 14, 2002 Kokubo et al.
20020038430 March 28, 2002 Edwards et al.
20020091819 July 11, 2002 Melchione et al.
20020095607 July 18, 2002 Lin-Hendel
20020116627 August 22, 2002 Tarbotton et al.
20020144156 October 3, 2002 Copeland
20020162015 October 31, 2002 Fang
20020166063 November 7, 2002 Lachman et al.
20020169952 November 14, 2002 DiSanto et al.
20020184528 December 5, 2002 Shevenell et al.
20020188887 December 12, 2002 Largman et al.
20020194490 December 19, 2002 Halperin et al.
20030021728 January 30, 2003 Sharpe et al.
20030074578 April 17, 2003 Ford et al.
20030084318 May 1, 2003 Schertz
20030101381 May 29, 2003 Mateev et al.
20030115483 June 19, 2003 Liang
20030188190 October 2, 2003 Aaron et al.
20030191957 October 9, 2003 Hypponen et al.
20030200460 October 23, 2003 Morota et al.
20030212902 November 13, 2003 van der Made
20030229801 December 11, 2003 Kouznetsov et al.
20030237000 December 25, 2003 Denton et al.
20040003323 January 1, 2004 Bennett et al.
20040006473 January 8, 2004 Mills et al.
20040015712 January 22, 2004 Szor
20040019832 January 29, 2004 Arnold et al.
20040047356 March 11, 2004 Bauer
20040083408 April 29, 2004 Spiegel et al.
20040088581 May 6, 2004 Brawn et al.
20040093513 May 13, 2004 Cantrell et al.
20040111531 June 10, 2004 Staniford et al.
20040117478 June 17, 2004 Triulzi et al.
20040117624 June 17, 2004 Brandt et al.
20040128355 July 1, 2004 Chao et al.
20040165588 August 26, 2004 Pandya
20040236963 November 25, 2004 Danford et al.
20040243349 December 2, 2004 Greifeneder et al.
20040249911 December 9, 2004 Alkhatib et al.
20040255161 December 16, 2004 Cavanaugh
20040268147 December 30, 2004 Wiederin et al.
20050005159 January 6, 2005 Oliphant
20050021740 January 27, 2005 Bar et al.
20050033960 February 10, 2005 Vialen et al.
20050033989 February 10, 2005 Poletto et al.
20050050148 March 3, 2005 Mohammadioun et al.
20050086523 April 21, 2005 Zimmer et al.
20050091513 April 28, 2005 Mitomo et al.
20050091533 April 28, 2005 Omote et al.
20050091652 April 28, 2005 Ross et al.
20050108562 May 19, 2005 Khazan et al.
20050114663 May 26, 2005 Cornell et al.
20050125195 June 9, 2005 Brendel
20050149726 July 7, 2005 Joshi et al.
20050157662 July 21, 2005 Bingham et al.
20050183143 August 18, 2005 Anderholm et al.
20050201297 September 15, 2005 Peikari
20050210533 September 22, 2005 Copeland et al.
20050238005 October 27, 2005 Chen et al.
20050240781 October 27, 2005 Gassoway
20050262562 November 24, 2005 Gassoway
20050265331 December 1, 2005 Stolfo
20050283839 December 22, 2005 Cowburn
20060010495 January 12, 2006 Cohen et al.
20060015416 January 19, 2006 Hoffman et al.
20060015715 January 19, 2006 Anderson
20060015747 January 19, 2006 Van de Ven
20060021029 January 26, 2006 Brickell et al.
20060021054 January 26, 2006 Costa et al.
20060031476 February 9, 2006 Mathes et al.
20060047665 March 2, 2006 Neil
20060070130 March 30, 2006 Costea et al.
20060075496 April 6, 2006 Carpenter et al.
20060095968 May 4, 2006 Portolani et al.
20060101516 May 11, 2006 Sudaharan et al.
20060101517 May 11, 2006 Banzhof et al.
20060117385 June 1, 2006 Mester et al.
20060123477 June 8, 2006 Raghavan et al.
20060143709 June 29, 2006 Brooks et al.
20060150249 July 6, 2006 Gassen et al.
20060161983 July 20, 2006 Cothrell et al.
20060161987 July 20, 2006 Levy-Yurista
20060161989 July 20, 2006 Reshef et al.
20060164199 July 27, 2006 Gilde et al.
20060173992 August 3, 2006 Weber et al.
20060179147 August 10, 2006 Tran et al.
20060184632 August 17, 2006 Marino et al.
20060191010 August 24, 2006 Benjamin
20060221956 October 5, 2006 Narayan et al.
20060236393 October 19, 2006 Kramer et al.
20060242709 October 26, 2006 Seinfeld et al.
20060248519 November 2, 2006 Jaeger et al.
20060248582 November 2, 2006 Panjwani et al.
20060251104 November 9, 2006 Koga
20060288417 December 21, 2006 Bookbinder et al.
20070006288 January 4, 2007 Mayfield et al.
20070006313 January 4, 2007 Porras et al.
20070011174 January 11, 2007 Takaragi et al.
20070016951 January 18, 2007 Piccard et al.
20070019286 January 25, 2007 Kikuchi
20070033645 February 8, 2007 Jones
20070038943 February 15, 2007 FitzGerald et al.
20070064689 March 22, 2007 Shin et al.
20070074169 March 29, 2007 Chess et al.
20070094730 April 26, 2007 Bhikkaji et al.
20070101435 May 3, 2007 Konanka et al.
20070128855 June 7, 2007 Cho et al.
20070142030 June 21, 2007 Sinha et al.
20070143827 June 21, 2007 Nicodemus et al.
20070156895 July 5, 2007 Vuong
20070157180 July 5, 2007 Tillmann et al.
20070157306 July 5, 2007 Elrod et al.
20070168988 July 19, 2007 Eisner et al.
20070171824 July 26, 2007 Ruello et al.
20070174915 July 26, 2007 Gribble et al.
20070192500 August 16, 2007 Lum
20070192858 August 16, 2007 Lum
20070198275 August 23, 2007 Malden et al.
20070208822 September 6, 2007 Wang et al.
20070220607 September 20, 2007 Sprosts et al.
20070240218 October 11, 2007 Tuvell et al.
20070240219 October 11, 2007 Fuvell et al.
20070240220 October 11, 2007 Fuvell et al.
20070240222 October 11, 2007 Fuvell et al.
20070250930 October 25, 2007 Aziz et al.
20070256132 November 1, 2007 Oliphant
20070271446 November 22, 2007 Nakamura
20080005782 January 3, 2008 Aziz
20080018122 January 24, 2008 Zierler et al.
20080028463 January 31, 2008 Dagon et al.
20080040710 February 14, 2008 Chiriac
20080046781 February 21, 2008 Childs et al.
20080066179 March 13, 2008 Liu
20080072326 March 20, 2008 Danford et al.
20080077793 March 27, 2008 Tan et al.
20080080518 April 3, 2008 Hoeflin et al.
20080086720 April 10, 2008 Lekel
20080098476 April 24, 2008 Syversen
20080120722 May 22, 2008 Sima et al.
20080134178 June 5, 2008 Fitzgerald et al.
20080134334 June 5, 2008 Kim et al.
20080141376 June 12, 2008 Clausen et al.
20080184367 July 31, 2008 McMillan et al.
20080184373 July 31, 2008 Traut et al.
20080189787 August 7, 2008 Arnold et al.
20080201778 August 21, 2008 Guo et al.
20080209557 August 28, 2008 Herley et al.
20080215742 September 4, 2008 Goldszmidt et al.
20080222729 September 11, 2008 Chen et al.
20080263665 October 23, 2008 Ma et al.
20080295172 November 27, 2008 Bohacek
20080301810 December 4, 2008 Lehane et al.
20080307524 December 11, 2008 Singh et al.
20080313738 December 18, 2008 Enderby
20080320594 December 25, 2008 Jiang
20090003317 January 1, 2009 Kasralikar et al.
20090007100 January 1, 2009 Field et al.
20090013408 January 8, 2009 Schipka
20090031423 January 29, 2009 Liu et al.
20090036111 February 5, 2009 Danford et al.
20090037835 February 5, 2009 Goldman
20090044024 February 12, 2009 Oberheide et al.
20090044274 February 12, 2009 Budko et al.
20090064332 March 5, 2009 Porras et al.
20090077666 March 19, 2009 Chen et al.
20090083369 March 26, 2009 Marmor
20090083855 March 26, 2009 Apap et al.
20090089879 April 2, 2009 Wang et al.
20090094697 April 9, 2009 Proves et al.
20090113425 April 30, 2009 Ports et al.
20090125976 May 14, 2009 Wassermann et al.
20090126015 May 14, 2009 Monastyrsky et al.
20090126016 May 14, 2009 Sobko et al.
20090133125 May 21, 2009 Choi et al.
20090144823 June 4, 2009 Lamastra et al.
20090158430 June 18, 2009 Borders
20090172815 July 2, 2009 Gu et al.
20090187992 July 23, 2009 Poston
20090193293 July 30, 2009 Stolfo et al.
20090198651 August 6, 2009 Shiffer et al.
20090198670 August 6, 2009 Shiffer et al.
20090198689 August 6, 2009 Frazier et al.
20090199274 August 6, 2009 Frazier et al.
20090199296 August 6, 2009 Xie et al.
20090228233 September 10, 2009 Anderson et al.
20090241187 September 24, 2009 Troyansky
20090241190 September 24, 2009 Todd et al.
20090265692 October 22, 2009 Godefroid et al.
20090271867 October 29, 2009 Zhang
20090300415 December 3, 2009 Zhang et al.
20090300761 December 3, 2009 Park et al.
20090328185 December 31, 2009 Berg et al.
20090328221 December 31, 2009 Blumfield et al.
20100005146 January 7, 2010 Drako et al.
20100011205 January 14, 2010 McKenna
20100017546 January 21, 2010 Poo et al.
20100030996 February 4, 2010 Butler, II
20100031353 February 4, 2010 Thomas et al.
20100037314 February 11, 2010 Perdisci et al.
20100043073 February 18, 2010 Kuwamura
20100054278 March 4, 2010 Stolfo et al.
20100058474 March 4, 2010 Hicks
20100064044 March 11, 2010 Nonoyama
20100077481 March 25, 2010 Polyakov et al.
20100083376 April 1, 2010 Pereira et al.
20100115621 May 6, 2010 Staniford et al.
20100132038 May 27, 2010 Zaitsev
20100154056 June 17, 2010 Smith et al.
20100180344 July 15, 2010 Malyshev et al.
20100192223 July 29, 2010 Ismael et al.
20100220863 September 2, 2010 Dupaquis et al.
20100235831 September 16, 2010 Dittmer
20100251104 September 30, 2010 Massand
20100281102 November 4, 2010 Chinta et al.
20100281541 November 4, 2010 Stolfo et al.
20100281542 November 4, 2010 Stolfo et al.
20100287260 November 11, 2010 Peterson et al.
20100299754 November 25, 2010 Amit et al.
20100306173 December 2, 2010 Frank
20110004737 January 6, 2011 Greenebaum
20110025504 February 3, 2011 Lyon et al.
20110041179 February 17, 2011 St Hlberg
20110047594 February 24, 2011 Mahaffey et al.
20110047620 February 24, 2011 Mahaffey et al.
20110055907 March 3, 2011 Narasimhan et al.
20110078794 March 31, 2011 Manni et al.
20110093951 April 21, 2011 Aziz
20110099620 April 28, 2011 Stavrou et al.
20110099633 April 28, 2011 Aziz
20110099635 April 28, 2011 Silberman et al.
20110113231 May 12, 2011 Kaminsky
20110145918 June 16, 2011 Jung et al.
20110145920 June 16, 2011 Mahaffey et al.
20110145934 June 16, 2011 Abramovici et al.
20110167493 July 7, 2011 Song et al.
20110167494 July 7, 2011 Bowen et al.
20110173213 July 14, 2011 Frazier et al.
20110173460 July 14, 2011 Ito et al.
20110219449 September 8, 2011 St. Neitzel et al.
20110219450 September 8, 2011 McDougal et al.
20110225624 September 15, 2011 Sawhney et al.
20110225655 September 15, 2011 Niemela et al.
20110247072 October 6, 2011 Staniford et al.
20110265182 October 27, 2011 Peinado et al.
20110289582 November 24, 2011 Kejriwal et al.
20110302587 December 8, 2011 Nishikawa et al.
20110307954 December 15, 2011 Melnik et al.
20110307955 December 15, 2011 Kaplan et al.
20110307956 December 15, 2011 Yermakov et al.
20110314546 December 22, 2011 Aziz et al.
20120023593 January 26, 2012 Puder et al.
20120054869 March 1, 2012 Yen et al.
20120066698 March 15, 2012 Yanoo
20120079596 March 29, 2012 Thomas et al.
20120084859 April 5, 2012 Radinsky et al.
20120096553 April 19, 2012 Srivastava et al.
20120110667 May 3, 2012 Zubrilin et al.
20120117652 May 10, 2012 Manni et al.
20120121154 May 17, 2012 Xue et al.
20120124426 May 17, 2012 Maybee et al.
20120174186 July 5, 2012 Aziz et al.
20120174196 July 5, 2012 Bhogavilli et al.
20120174218 July 5, 2012 McCoy et al.
20120198279 August 2, 2012 Schroeder
20120210423 August 16, 2012 Friedrichs et al.
20120222121 August 30, 2012 Staniford et al.
20120255015 October 4, 2012 Sahita et al.
20120255017 October 4, 2012 Sallam
20120260342 October 11, 2012 Dube et al.
20120266244 October 18, 2012 Green et al.
20120278886 November 1, 2012 Luna
20120297489 November 22, 2012 Dequevy
20120330801 December 27, 2012 McDougal et al.
20120331553 December 27, 2012 Aziz et al.
20130014259 January 10, 2013 Gribble et al.
20130036472 February 7, 2013 Aziz
20130047257 February 21, 2013 Aziz
20130074185 March 21, 2013 McDougal et al.
20130086684 April 4, 2013 Mohler
20130097699 April 18, 2013 Balupar et al.
20130097706 April 18, 2013 Titonis et al.
20130111587 May 2, 2013 Goel et al.
20130117852 May 9, 2013 Stute
20130117855 May 9, 2013 Kim et al.
20130139264 May 30, 2013 Brinkley et al.
20130160125 June 20, 2013 Likhachev et al.
20130160127 June 20, 2013 Jeong et al.
20130160130 June 20, 2013 Mendelev et al.
20130160131 June 20, 2013 Madou et al.
20130167236 June 27, 2013 Sick
20130174214 July 4, 2013 Duncan
20130185789 July 18, 2013 Hagiwara et al.
20130185795 July 18, 2013 Winn et al.
20130185798 July 18, 2013 Saunders et al.
20130191915 July 25, 2013 Antonakakis et al.
20130196649 August 1, 2013 Paddon et al.
20130227691 August 29, 2013 Aziz et al.
20130246370 September 19, 2013 Bartram et al.
20130247186 September 19, 2013 LeMasters
20130263260 October 3, 2013 Mahaffey et al.
20130291109 October 31, 2013 Staniford et al.
20130298243 November 7, 2013 Kumar et al.
20130318038 November 28, 2013 Shiffer et al.
20130318073 November 28, 2013 Shiffer et al.
20130325791 December 5, 2013 Shiffer et al.
20130325792 December 5, 2013 Shiffer et al.
20130325871 December 5, 2013 Shiffer et al.
20130325872 December 5, 2013 Shiffer et al.
20140032875 January 30, 2014 Butler
20140053260 February 20, 2014 Gupta et al.
20140053261 February 20, 2014 Gupta et al.
20140130158 May 8, 2014 Wang et al.
20140137180 May 15, 2014 Lukacs et al.
20140169762 June 19, 2014 Ryu
20140179360 June 26, 2014 Jackson et al.
20140181131 June 26, 2014 Ross
20140189687 July 3, 2014 Jung et al.
20140189866 July 3, 2014 Shiffer et al.
20140189882 July 3, 2014 Jung et al.
20140237600 August 21, 2014 Silberman et al.
20140280245 September 18, 2014 Wilson
20140283037 September 18, 2014 Sikorski et al.
20140283063 September 18, 2014 Thompson et al.
20140328204 November 6, 2014 Klotsche et al.
20140337836 November 13, 2014 Ismael
20140344926 November 20, 2014 Cunningham et al.
20140351930 November 27, 2014 Sun
20140351935 November 27, 2014 Shao et al.
20140380473 December 25, 2014 Bu et al.
20140380474 December 25, 2014 Paithane et al.
20150007312 January 1, 2015 Pidathala et al.
20150096022 April 2, 2015 Vincent et al.
20150096023 April 2, 2015 Mesdaq et al.
20150096024 April 2, 2015 Haq et al.
20150096025 April 2, 2015 Ismael
20150180886 June 25, 2015 Staniford et al.
20150186645 July 2, 2015 Aziz et al.
20150199513 July 16, 2015 Ismael et al.
20150199531 July 16, 2015 Ismael et al.
20150199532 July 16, 2015 Ismael et al.
20150220735 August 6, 2015 Paithane et al.
20150372980 December 24, 2015 Eyada
20160004869 January 7, 2016 Ismael et al.
20160006756 January 7, 2016 Ismael et al.
20160044000 February 11, 2016 Cunningham
20160127393 May 5, 2016 Aziz et al.
20160191547 June 30, 2016 Zafar et al.
20160191550 June 30, 2016 Ismael et al.
20160261612 September 8, 2016 Mesdaq et al.
20160285914 September 29, 2016 Singh et al.
20160301703 October 13, 2016 Aziz
20160335110 November 17, 2016 Paithane et al.
20170083703 March 23, 2017 Abbasi et al.
20180013770 January 11, 2018 Ismael
20180048660 February 15, 2018 Paithane et al.
20180121316 May 3, 2018 Ismael et al.
20180288077 October 4, 2018 Siddiqui et al.
20190268152 August 29, 2019 Sandoval
Foreign Patent Documents
2439806 January 2008 GB
2490431 October 2012 GB
0206928 January 2002 WO
02/23805 March 2002 WO
2007117636 October 2007 WO
2008/041950 April 2008 WO
2011/084431 July 2011 WO
2011/112348 September 2011 WO
2012/075336 June 2012 WO
2012/145066 October 2012 WO
2013/067505 May 2013 WO
Other references
  • Venezia, Paul, “NetDetector Captures Intrusions”, InfoWorld Issue 27, (“Venezia”), (Jul. 14, 2003).
  • Vladimir Getov: “Security as a Service in Smart Clouds—Opportunities and Concerns”, Computer Software and Applications Conference (COMPSAC), 2012 IEEE 36th Annual, IEEE, Jul. 16, 2012 (Jul. 16, 2012).
  • Wahid et al., Characterising the Evolution in Scanning Activity of Suspicious Hosts, Oct. 2009, Third International Conference on Network and System Security, pp. 344-350.
  • Whyte, et al., “DNS-Based Detection of Scanning Works in an Enterprise Network”, Proceedings of the 12th Annual Network and Distributed System Security Symposium, (Feb. 2005), 15 pages.
  • Williamson, Matthew M., “Throttling Viruses: Restricting Propagation to Defeat Malicious Mobile Code”, ACSAC Conference, Las Vegas, NV, USA, (Dec. 2002), pp. 1-9.
  • Yuhei Kawakoya et al: “Memory behavior-based automatic malware unpacking in stealth debugging environment”, Malicious and Unwanted Software (Malware), 2010 5th International Conference on, IEEE, Piscataway, NJ, USA, Oct. 19, 2010, pp. 39-46, XP031833827, ISBN:978-1-4244-8-9353-1.
  • Zhang et al., The Effects of Threading, Infection Time, and Multiple-Attacker Collaboration on Malware Propagation, Sep. 2009, IEEE 28th International Symposium on Reliable Distributed Systems, pp. 73-82.
  • “Mining Specification of Malicious Behavior”—Jha et al., UCSB, Sep. 2007 https://www.cs.ucsb.edu/.about.chris/research/doc/esec07.sub.-mining.pdf-.
  • “Network Security: NetDetector—Network Intrusion Forensic System (NIFS) Whitepaper”, (“NetDetector Whitepaper”), (2003).
  • “When Virtual is Better Than Real”, IEEEXplore Digital Library, available at, http://ieeexplore.ieee.org/xpl/articleDetails.isp?reload=true&arnumbe- r=990073, (Dec. 7, 2013).
  • Abdullah, et al., Visualizing Network Data for Intrusion Detection, 2005 IEEE Workshop on Information Assurance and Security, pp. 100-108.
  • Adetoye, Adedayo , et al., “Network Intrusion Detection & Response System”, (“Adetoye”), (Sep. 2003).
  • Apostolopoulos, George; hassapis, Constantinos; “V-eM: A cluster of Virtual Machines for Robust, Detailed, and High-Performance Network Emulation”, 14th IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, Sep. 11-14, 2006, pp. 117-126.
  • Aura, Tuomas, “Scanning electronic documents for personally identifiable information”, Proceedings of the 5th ACM workshop on Privacy in electronic society. ACM, 2006.
  • Baecher, “The Nepenthes Platform: An Efficient Approach to collect Malware”, Springer-verlag Berlin Heidelberg, (2006), pp. 165-184.
  • Bayer, et al., “Dynamic Analysis of Malicious Code”, J Comput Virol, Springer-Verlag, France., (2006), pp. 67-77.
  • Boubalos, Chris , “extracting syslog data out of raw pcap dumps, seclists.org, Honeypots mailing list archives”, available at http://seclists.org/honeypots/2003/q2/319 (“Boubalos”), (Jun. 5, 2003).
  • Chaudet, C., et al., “Optimal Positioning of Active and Passive Monitoring Devices”, International Conference on Emerging Networking Experiments and Technologies, Proceedings of the 2005 ACM Conference on Emerging Network Experiment and Technology, CoNEXT '05, Toulousse, France, (Oct. 2005), pp. 71-82.
  • Chen, P. M. and Noble, B. D., “When Virtual is Better Than Real, Department of Electrical Engineering and Computer Science”, University of Michigan (“Chen”) (2001).
  • Cisco “Intrusion Prevention for the Cisco ASA 5500-x Series” Data Sheet (2012).
  • Cohen, M.I. , “PyFlag—An advanced network forensic framework”, Digital investigation 5, Elsevier, (2008), pp. S112-5120.
  • Costa, M. , et al., “Vigilante: End-to-End Containment of Internet Worms”, SOSP '05, Association for Computing Machinery, Inc., Brighton U.K., (Oct. 23-26, 2005).
  • Didier Stevens, “Malicious PDF Documents Explained”, Security & Privacy, IEEE, IEEE Service Center, Los Alamitos, CA, US, vol. 9, No. 1, Jan. 1, 2011, pp. 80-82, XP011329453, Issn: 1540-7993, DOI: 10.1109/MSP.2011.14.
  • Distler, “Malware Analysis: An Introduction”, SANS Institute InfoSec Reading Room, SANS Institute, (2007).
  • Dunlap, George W., et al., “ReVirt: Enabling Intrusion Analysis through Virtual-Machine Logging and Replay”, Proceeding of the 5th Symposium on Operating Systems Design and Implementation, USENIX Association, (“Dunlap”), (Dec. 9, 2002).
  • FireEye Malware Analysis & Exchange Network, Malware Protection System, FireEye Inc., 2010.
  • FireEye Malware Analysis, Modern Malware Forensics, FireEye Inc., 2010.
  • FireEye v.6.0 Security Target, pp. 1-35, Version 1.1, FireEye Inc., May 2011.
  • Goel, et al., Reconstructing System State for Intrusion Analysis, Apr. 2008 SIGOPS Operating Systems Review, vol. 42 Issue 3, pp. 21-28.
  • Gregg Keizer: “Microsoft's HoneyMonkeys Show Patching Windows Works”, Aug. 8, 2005, XP055143386, Retrieved from the Internet: URL:http://www.informationweek.com/microsofts-honeymonkeys-show-patching-windows-works/d/d-id/1035069? [retrieved on Jun. 1, 2016].
  • Heng Yin et al, Panorama: Capturing System-Wide Information Flow for Malware Detection and Analysis, Research Showcase @ CMU, Carnegie Mellon University, 2007.
  • Hiroshi Shinotsuka, Malware Authors Using New Techniques to Evade Automated Threat Analysis Systems, Oct. 26, 2012, http://www.symantec.com/connect/blogs/, pp. 1-4.
  • Idika et al., A-Survey-of-Malware-Detection-Techniques, Feb. 2, 2007, Department of Computer Science, Purdue University.
  • Isohara, Takamasa, Keisuke Takemori, and Ayumu Kubota. “Kernel-based behavior analysis for android malware detection.” Computational intelligence and Security (CIS), 2011 Seventh International Conference on. IEEE, 2011.
  • Kaeo, Merike , “Designing Network Security”, (“Kaeo”), (Nov. 2003).
  • Kevin A Roundy et al: “Hybrid Analysis and Control of Malware”, Sep. 15, 2010, Recent Advances in Intrusion Detection, Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 317-338, XP019150454 ISBN:978-3-642-15511-6.
  • Khaled Salah et al: “Using Cloud Computing to Implement a Security Overlay Network”, Security & Privacy, IEEE, IEEE Service Center, Los Alamitos, CA, US, vol. 11, No. 1, Jan. 1, 2013 (Jan. 1, 2013).
  • Kim, H. , et al., “Autograph: Toward Automated, Distributed Worm Signature Detection”, Proceedings of the 13th Usenix Security Symposium (Security 2004), San Diego, (Aug. 2004), pp. 271-286.
  • King, Samuel T., et al., “Operating System Support for Virtual Machines”, (“King”), (2003).
  • Kreibich, C. , et al., “Honeycomb-Creating Intrusion Detection Signatures Using Honeypots”, 2nd Workshop on Hot Topics in Networks (HotNets-11), Boston, USA, (2003).
  • Kristoff, J. , “Botnets, Detection and Mitigation: DNS-Based Techniques”, NU Security Day, (2005), 23 pages.
  • Lastline Labs, The Threat of Evasive Malware, Feb. 25, 2013, Lastline Labs, pp. 1-8.
  • Li et al., A VMM-Based System Call Interposition Framework for Program Monitoring, Dec. 2010, IEEE 16th International Conference on Parallel and Distributed Systems, pp. 706-711.
  • Lindorfer, Martina, Clemens Kolbitsch, and Paolo Milani Comparetti. “Detecting environment-sensitive malware.” Recent Advances in Intrusion Detection. Springer Berlin Heidelberg, 2011.
  • Marchette, David J., “Computer Intrusion Detection and Network Monitoring: A Statistical Viewpoint”, (“Marchette”), (2001).
  • Moore, D., et al., “Internet Quarantine: Requirements for Containing Self-Propagating Code”, INFOCOM, vol. 3, (Mar. 30-Apr. 3, 2003), pp. 1901-1910.
  • Morales, Jose A., et al., ““Analyzing and exploiting network behaviors of malware.””, Security and Privacy in Communication Networks. Springer Berlin Heidelberg, 2010. 20-34.
  • Mori, Detecting Unknown Computer Viruses, 2004, Springer-Verlag Berlin Heidelberg.
  • Natvig, Kurt, “SANDBOXII: Internet”, Virus Bulletin Conference, (“Natvig”), (Sep. 2002).
  • NetBIOS Working Group. Protocol Standard fora NetBIOS Service on a TCP/UDP transport: Concepts and Methods. STD 19, RFC 1001, Mar. 1987.
  • Newsome, J. , et al., “Dynamic Taint Analysis for Automatic Detection, Analysis, and Signature Generation of Exploits on Commodity Software”, In Proceedings of the 12th Annual Network and Distributed System Security, Symposium (NDSS '05), (Feb. 2005).
  • Nojiri, D. , et al., “Cooperation Response Strategies for Large Scale Attack Mitigation”, DARPA Information Survivability Conference and Exposition, vol. 1, (Apr. 22-24, 2003), pp. 293-302.
  • Oberheide et al., CloudAV.sub.--N-Version Antivirus in the Network Cloud, 17th USENIX Security Symposium USENIX Security '8 Jul. 28-Aug. 1, 2008 San Jose, CA.
  • Reiner Sailer, Enriquillo Valdez, Trent Jaeger, Roonald Perez, Leendert van Doom, John Linwood Griffin, Stefan Berger., sHype: Secure Hypervisor Appraoch to Trusted Virtualized Systems (Feb. 2, 2005) (“Sailer”).
  • Silicon Defense, “Worm Containment in the Internal Network”, (Mar. 2003), pp. 1-25.
  • Singh, S. , et al., “Automated Worm Fingerprinting”, Proceedings of the ACM/USENIX Symposium on Operating System Design and Implementation, San Francisco, California, (Dec. 2004).
  • Thomas H. Ptacek, and Timothy N. Newsham , “Insertion, Evasion, and Denial of Service: Eluding Network Intrusion Detection”, Secure Networks, (“Ptacek”), (Jan. 1998).
Patent History
Patent number: 11314859
Type: Grant
Filed: Mar 14, 2019
Date of Patent: Apr 26, 2022
Assignee: FireEye Security Holdings, Inc. (Milpitas, CA)
Inventors: Japneet Singh (Bangalore), Ratnesh Pandey (Allahabad), Atul Kabra (Bangalore)
Primary Examiner: Jeffrey C Pwu
Assistant Examiner: Nega Woldemariam
Application Number: 16/353,984
Classifications
Current U.S. Class: Monitoring Or Scanning Of Software Or Data Including Attack Prevention (726/22)
International Classification: H04L 9/32 (20060101); G06F 21/55 (20130101); G06F 21/56 (20130101); G06F 21/60 (20130101); H04L 29/06 (20060101); G06F 21/62 (20130101);